Journal of Infectious Diseases Advance Access published September 3, 2015
MAJOR ARTICLE
Genes Required for Free Phage Production are Essential for Pseudomonas aeruginosa Chronic Lung Infections Andrée-Ann Lemieux,1,a Julie Jeukens,1,a Irena Kukavica-Ibrulj,1,a Joanne L. Fothergill,2,a Brian Boyle,1 Jérôme Laroche,1 Nicholas P. Tucker,3 Craig Winstanley,2 and Roger C. Levesque1 1
Institut de biologie intégrative et des systèmes, Université Laval, Québec City, Canada; 2Institute of Infection and Global Health, University of Liverpool, and 3Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, Glasgow, United Kingdom
The opportunistic pathogen Pseudomonas aeruginosa causes chronic lung infection in patients with cystic fibrosis. The Liverpool Epidemic Strain LESB58 is highly resistant to antibiotics, transmissible, and associated with increased morbidity and mortality. Its genome contains 6 prophages and 5 genomic islands. We constructed a polymerase chain reaction (PCR)–based signature-tagged mutagenesis library of 9216 LESB58 mutants and screened the mutants in a rat model of chronic lung infection. A total of 162 mutants were identified as defective for in vivo maintenance, with 11 signature-tagged mutagenesis mutants having insertions in prophage and genomic island genes. Many of these mutants showed both diminished virulence and reduced phage production. Transcription profiling by quantitative PCR and RNA-Seq suggested that disruption of these prophages had a widespread trans-acting effect on the transcriptome. This study demonstrates that temperate phages play a pivotal role in the establishment of infection through modulation of bacterial host gene expression. Keywords. expression.
Pseudomonas aeruginosa; bacteriophages; virulence; chronic infection; respiratory health; gene
Pseudomonas aeruginosa is an opportunistic pathogen responsible for lung disease in immunocompromised and patients with cystic fibrosis (CF) [1]. The widespread assumption that patients with CF acquire only unique strains of P. aeruginosa from the environment was challenged when molecular typing was used to demonstrate the spread of a β-lactam–resistant isolate, now known as the Liverpool epidemic strain (LES). The LES was first identified in a United Kingdom children’s CF unit in 1996 [2] but has since been identified
Received 8 June 2015; accepted 31 July 2015. Presented in part: 13th International Conference on Pseudomonas, Sydney, Australia, 4th–7th September 2011; European Cystic Fibrosis Society Conference, Hamburg, Germany, 8th–11th June 2011. a A.-A. L., J. J., I. K.-I., and J. L. F. contributed equally to this work. Correspondence: Roger C. Levesque, PhD, Institut de biologie intégrative et des systèmes (IBIS), Université Laval, 1030 avenue de la médecine, Québec City, QC G1V 0A6, Canada (
[email protected]). The Journal of Infectious Diseases® © The Author 2015. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail:
[email protected]. DOI: 10.1093/infdis/jiv415
in North America [3–5]. Some LES isolates, including the fully sequenced isolate LESB58 [6], exhibit an unusual phenotype characterized by early overexpression of the cell-density–dependent quorum sensing (QS) regulon, including virulence-related secreted factors such as LasA, elastase, and pyocyanin [7–9]. The LES is associated with greater patient morbidity and mortality, compared with other P. aeruginosa strains [10], and was reported in unexpected cases of transmission from children with CF to their parents [11, 12]. Whole-genome sequencing of LESB58 revealed that 90% of the genome material was highly conserved and that a unique accessory genome of 455 genes was located within prophages (PPs) and genomic islands (GIs). Four of the 6 PPs are absent from the genome of strain PAO1, and only 2 of the 5 GIs identified in the LES showed similarity to any previously identified P. aeruginosa GI [6]. A previous study on LESB58 phage production indicated that 5 of the 6 PPs are able to produce active phage progeny [6]. These phages can be produced spontaneously or following a stress Phage Alters Virulence of P. aeruginosa
•
JID
•
1
response, such as induction with antibiotics [13]. The role of PPs in virulence and for in vivo initiation and maintenance of infection remains poorly understood [6]. In this report, we build upon these studies to investigate the importance of specific genes implicated in in vivo maintenance, especially those found in the accessory genome of LESB58. We performed polymerase chain reaction (PCR)–based signaturetagged mutagenesis (STM), a negative selection technique that allows the identification of genes assumed to be essential in vivo. Preliminary results for LESB58 STM and identification of 48 mutants have been published previously [6]. Here, we focus on the functional impact of these accessory genome mutations, with the goal of achieving a better understanding of the role played by mobile genetic elements in the maintenance of chronic lung infection among patients with CF. Wild-type strain LESB58 was compared with mutant strains for phenotypic characteristics, phage production, and transcriptome profiling, demonstrating a clear role for temperate phages in the infection process and modulation of bacterial host gene expression. MATERIALS AND METHODS Ethics Statement
The use of animals for this study was reviewed by the University Laval Committee for Animal Care ( protocol number 2011194). Bacterial Strains, Plasmids, Media, and Culture Conditions
P. aeruginosa LESB58 [6] and Escherichia coli were grown in tryptic soy broth (Difco) unless otherwise indicated. When needed, these media were supplemented with 1.5% Bacto-agar (Difco), ampicillin (100 µg/mL for Escherichia coli DH10B), and tetracycline (10 µg/mL for E. coli DH10B or 45 µg/mL for P. aeruginosa LESB58 STM mutants; Sigma-Aldrich). Cloning of chromosomal DNA was performed using the pTZ18R vector (GE Healthcare). Restriction enzymes, T4 DNA ligase, and T4 DNA polymerase were purchased from New England Biolabs. PCR-Based STM of LESB58
PCR-based STM is a well-defined method that has been used with P. aeruginosa PAO1 and adapted to strain LESB58 [32]. The rat model of chronic lung infection was used to screen 9216 mutants in vivo [33]. Disrupted genes were identified by Sanger sequencing as previously described [33] and confirmed by similarity searches with the P. aeruginosa LESB58 genome database (available at: http://www.pseudomonas.com). In Vitro and In Vivo Competitive Index
Agar beads were prepared according to a modified version of a previously described method [8]. From overnight cultures of each LESB58 STM mutant and the wild-type strain, a fresh culture was grown until an OD600 of 1 was reached. Subsequent 2
•
JID
•
Lemieux et al
steps were as described previously [8]. An input ratio of 1 (LESB58) to 1 (STM mutant) was used. Six animals were originally used for each strain. Owing to unsuccessful infection or mortality, the experiment was performed a second time for some of the strains. In vitro and in vivo competitive index (CI) analyses were performed according to modified versions of previously described methods [8]. Each in vivo and in vitro competition was tested for statistical significance, using the Mann–Whitney test, in GraphPrism Pro 5.0 software. Phenotypic Analysis of Mutant Strains
P. aeruginosa STM/PALES_08021 and STM/PALES_13521 strains were tested for biofilm formation, swimming, swarming and twitching motility, heat shock, hydrogen peroxidase, hemolysin, pyocyanin, pyoverdine, and protease assays as described previously [4, 6, 29], with the LESB58 strain used as a control. Spontaneous and Induced Phage Production
For spontaneous phage production, cultures were grown to an OD of 0.5 (A600nm). Phage levels were determined by incubation of cultures for 1 hour (without antibiotics), followed by a 1-hour recovery period in fresh medium. Induced active phage particles were enumerated using plaque assays, in which 100 µL of culture supernatant was added to 100 µL of 0.5 (A600nm) P. aeruginosa PAO1 (the indicator strain) in 5 mL of molten 0.4% (w/v) Luria agar and poured onto L-agar, in triplicate. It was previously demonstrated that plaques visible in this assay are produced by prophages 2 and 3 [13]. Phage induction using norfloxacin was performed by adding the antibiotic at the minimum inhibitory concentration (50 µg/mL). RNA Extraction, Complementary DNA (cDNA) Synthesis, and Quantitative PCR (qPCR)
Total RNA was extracted from bacterial cultures at mid-log phase that had an optical density of 0.6–0.8 at 650 nm, using the RNeasy Midi Kit (Invitrogen) according to the instructions of the supplier. cDNA was synthesized using 1 µg of total RNA as the template (Quantitect Reverse Transcription Kit [Qiagen] and random primers [Invitrogen]) according to the manufacturer’s protocol. qPCR was performed on the Roche Light Cycler 480. Primers used are shown in Supplementary Table 2. Differential gene expression between the wild-type and STM mutants was assessed using the Mann–Whitney test in GraphPad Prism 5.0. Transcriptome Profiling
Total RNA was depleted of ribosomal RNA and transfer RNA by using the RiboMinus Kit (Invitrogen). Total RNA and messenger RNA quality were assessed using the Agilent RNA 6000 Pico Kit and the BioAnalyzer 2010 (Agilent Technologies). cDNA synthesis was performed using the cDNA synthesis system from Roche according to the instructions of the supplier. cDNA samples were sent to the genomic platform at the Institut
de biologie integrative et des systems (Université Laval, Québec, Canada) for transcriptome sequencing, using the Roche 454 pyrosequencing method on the Genome Sequencer FLX system with Titanium chemistry. Data obtained from whole-transcriptome sequencing was assembled and analyzed with Newbler (Roche). Reads were mapped on the P. aeruginosa LESB58 reference genome available in the Pseudomonas genome database (available at: http://www.pseudomonas.com). Differential gene expression was assessed using the R BioConductor package EdgeR [34]. Statistical analysis was based on the negative binomial distribution that takes into account the presence of overdispersion across the samples. Three wild-type RNA-seq libraries were compared to 1 mutant library. Raw counts were directly used in the R package, and normalization of the number of reads between samples was done with the quantile-adjusted method. Separate dispersions were estimated for individual tags (tagwise dispersion approach). P values adjusted according to the method of Benjamini and Hochberg [35] were used to select differentially expressed genes (adjusted P < .01).
Identification of LESB58 Mutants Defective for In Vivo Maintenance
Figure 1. In vivo competitive index of mutant Pseudomonas aeruginosa LESB58 strains with an insertion in a prophage or genomic island. Mutants were grown in a rat lung in competition with wild-type (WT) strain LESB58 for 7 days. Each circle represents the survival ratio, or competitive index (CI), of the mutant to WT strains for a single animal. The geometric mean of the CI replicates for a mutant is shown as a solid line and corresponds to the CI indicated on the x-axis. All CIs were significantly
